Adapt-N: A Cloud Computational Tool for

Precision Nitrogen Management

Chris GrahamHarold van Es

Changing Paradigm:Nitrogen Managementwith Cloud Computational Tools• Universal access through web-based services• Move from generalized to adaptive, real-time, site-

specific recommendations using information on- Weather- Local soils - Soil and crop management

• Universal process-based approach- Incorporates system complexity, variability and risk

• Real-time and post-season evaluations• Progressive improvement and refinement

2

Adapt-N History• 1970’s through early 2000’s: field research on

space-time aspects of N response to fertilizer (response trials, lysimeter studies, etc.)

• 1980’s to early 2000’s: model development (LEACHM, Sinclair model, combined into PNM model)

• 2004-2007: Semi-weekly adaptive N recommendations through email

• 2008-2013: Adapt-N tool available through web-based user interface

• 2011-2014: extensive field testing through on-farm trials

• 2014: Adapt-N licensed and commercialized through Agronomic Technology Corporation

What factors does Adapt-N include in making a recommendation?

• High resolution daily precipitation & temperature data• Soils: texture/soil type, slope, rooting depth, % organic matter

(NRCS databases)• N fertilizer applications: rate, type, timing, placement• Cultivar: Silage, grain, or sweet corn; planting date, maturity class• Population and expected yield• Tillage: fall or spring plowing; conservation tillage/residue

management• Manure applications: date, rate, N analysis, incorporation info• Rotations: soy, corn - silage or grain, or sod - last 3 yrs, % legume,

surface killed or incorporated• Irrigation amounts and dates• Fertilizer and grain prices & profit loss risk

5

High Resolution Climate Data (4x4 km)Critical Input to Adapt-N Tool

Iowa, June 14, 2011

• Public-private partnership, independent from product sales

• Scientific integrity with continued advancements• Integration with complementary technologies (farm

GIS, data mgmt, field application softwares)• Better organization-level functionality and

customization• Commercial data center (bigger Cloud)

Adapt-N Commercialization through Agronomic Technology Corporation

Adapt-N.com

Adapt-N InterfaceDefining your location

Adapt-N InterfaceViewing your locations

Adapt-N InterfaceEntering Soil Information

Adapt-N InterfaceEntering Soils Information

Adapt-N InterfaceEntering Crop Information

Adapt-N InterfaceEntering N Fertilizer Inputs

Adapt-N InterfaceEntering Manure Inputs

Adapt-N InterfaceEntering Irrigation Inputs

Zone RecommendationDate selection

Farm Summary

Agronomist Summary

Standard Report - PDF

Daily Alerts

Text

21

Adapt-N GraphsPrecipitation and Total N Loss

22

Adapt-N Graphscumulative precipitation; leaching and gaseous N losses

23

Adapt-N GraphsMineralized N and Temperature

24

Adapt-N GraphsNitrogen Availability and Uptake

25

Adapt-N GraphsCrop Growth and N Uptake

2011 – 2013 Adapt-N On-Farm Trials

Over 100 replicated trials in 10 states. • Adapt-N vs. Grower rate: side-by-side; some with zero-N• N response trials (4-6 N rates)

Overall Adapt-N Performance2011-2013, New York and Iowa

Average Change due to Adapt-N use

(Adapt-N - Grower-N)

By State By N rate changeGrand MeanNY trials IA trials N decrease

(A<G)N increase

(A>G)

n=67 n=37 n=87 n=17 n=104

Total N fertilizer applied (lb/ac) -52 -29 -60 38 -44

Simulated N leaching loss (lb/ac)* -11 -1 -10 3 -8

Simulated N total loss (lb/ac)* -36 -4 -34 16 -26

Yield (bu/ac equivalent) 2 0 -2 17 1

Profit ($/ac) $37 $17 $23 $65 $30

• Nitrous oxide losses (NIFA-AFRI project)• Cover crops and soil health (NIFA-SARE project)• Integration with farm GIS and data management

software• Enhanced efficiency

compounds• Canopy sensors?

Adapt-N: In-progress

Image by Raven-Slingshot

Adapt-N Active Zones - July 2014

Bottom Line:What are Adapt-N’s Benefits?

• Profit increases, with overall less N use• Economic and environmental benefits• Transparency and insights• Incentives for better management

AcknowledgementsAdapt-N Core TeamHarold van EsJeff MelkonianArt DeGaetanoLaura JosephBianca Moebius-CluneBob Schindelbeck

MGT Envirotec – IowaAgronomic Technology Corporation

FundersUSDA-NIFA Special GrantUSDA-AFRINorthern New York Agricultural

Development ProgramNew York Farm Viability InstituteUSDA-NRCS McKnight, Packard and Walton

Family FoundationsEnvironmental Defense FundInternational Plant Nutrition

Institute

N Recommendation Methodology: deterministic-stochastic mass balance

 SidedressNrate = CropNHarvest - CropNCurrent - SoilNCurrent - SoilNpostsidedress -

SoybeanNcredit + NLosspostapplication - Correctprofit

Probabilistic simulations

Input: Expected Yield

Near-Real-Time Simulation at Sidedress

simulated & partial fixed credit